• DocumentCode
    1203936
  • Title

    Attractor-based trust-region algorithm for efficient training of multilayer perceptrons

  • Author

    Lee, Jaewook

  • Author_Institution
    Dept. of Ind. Eng., Pohang Univ. of Sci. & Technol., South Korea
  • Volume
    39
  • Issue
    9
  • fYear
    2003
  • fDate
    5/1/2003 12:00:00 AM
  • Firstpage
    727
  • Lastpage
    728
  • Abstract
    A new training algorithm for multilayer perceptrons is proposed. The first phase is a trust region-based local search for fast training of networks. The second phase is an attractor-based global search for escaping local minima. Benchmark results show that the proposed algorithm outperforms previously reported existing techniques.
  • Keywords
    learning (artificial intelligence); multilayer perceptrons; MLPs; attractor-based global search; attractor-based trust-region algorithm; fast training; local minima escape; multilayer perceptron training algorithm; supervised learning method; trust region-based local search;
  • fLanguage
    English
  • Journal_Title
    Electronics Letters
  • Publisher
    iet
  • ISSN
    0013-5194
  • Type

    jour

  • DOI
    10.1049/el:20030498
  • Filename
    1199962
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=1203936